diff --git a/NEWS b/NEWS index a021293adc163..cc197c741b30a 100644 --- a/NEWS +++ b/NEWS @@ -25,6 +25,9 @@ PHP NEWS Moreover, some ext/date exceptions have been wrapped inside a IntlException now. (Girgias) . The intl.error_level INI setting has been deprecated. (Girgias) + . Added grapheme_strpos(), grapheme_stripos(), grapheme_strrpos(), + grapheme_strripos(), grapheme_strstr(), grapheme_stristr() and + grapheme_levenshtein() functions add $locale parameter (Yuya Hamada). - MySQLi: . The mysqli_execute() alias function has been deprecated. (timwolla) diff --git a/UPGRADING b/UPGRADING index 1d5b2b8ede499..baae19bf07b14 100644 --- a/UPGRADING +++ b/UPGRADING @@ -407,7 +407,10 @@ PHP 8.5 UPGRADE NOTES TransLiterator::getErrorCode(), and TransLiterator::getErrorMessage() have dropped the false from the return type union. Returning false was actually never possible. - + . grapheme_strpos(), grapheme_stripos(), grapheme_strrpos(), + grapheme_strripos(), grapheme_strstr(), grapheme_stristr() and + grapheme_levenshtein() functions add $locale parameter. + RFC: https://wiki.php.net/rfc/grapheme_add_locale_for_case_insensitive - LDAP: . ldap_get_option() now accepts a NULL connection, as ldap_set_option(), to allow retrieval of global options. diff --git a/ext/intl/grapheme/grapheme_string.c b/ext/intl/grapheme/grapheme_string.c index 8abef7dd560aa..023512cea08d8 100644 --- a/ext/intl/grapheme/grapheme_string.c +++ b/ext/intl/grapheme/grapheme_string.c @@ -81,19 +81,20 @@ PHP_FUNCTION(grapheme_strlen) /* {{{ Find position of first occurrence of a string within another */ PHP_FUNCTION(grapheme_strpos) { - char *haystack, *needle; - size_t haystack_len, needle_len; + char *haystack, *needle, *locale = ""; + size_t haystack_len, needle_len, locale_len = 0; const char *found; zend_long loffset = 0; int32_t offset = 0; size_t noffset = 0; zend_long ret_pos; - ZEND_PARSE_PARAMETERS_START(2, 3) + ZEND_PARSE_PARAMETERS_START(2, 4) Z_PARAM_STRING(haystack, haystack_len) Z_PARAM_STRING(needle, needle_len) Z_PARAM_OPTIONAL Z_PARAM_LONG(loffset) + Z_PARAM_STRING(locale, locale_len) ZEND_PARSE_PARAMETERS_END(); if ( OUTSIDE_STRING(loffset, haystack_len) ) { @@ -121,7 +122,7 @@ PHP_FUNCTION(grapheme_strpos) } /* do utf16 part of the strpos */ - ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, NULL, 0 /* fIgnoreCase */, 0 /* last */ ); + ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, NULL, 0 /* fIgnoreCase */, 0, locale /* last */ ); if ( ret_pos >= 0 ) { RETURN_LONG(ret_pos); @@ -134,19 +135,20 @@ PHP_FUNCTION(grapheme_strpos) /* {{{ Find position of first occurrence of a string within another, ignoring case differences */ PHP_FUNCTION(grapheme_stripos) { - char *haystack, *needle; - size_t haystack_len, needle_len; + char *haystack, *needle, *locale = ""; + size_t haystack_len, needle_len, locale_len = 0; const char *found; zend_long loffset = 0; int32_t offset = 0; zend_long ret_pos; int is_ascii; - ZEND_PARSE_PARAMETERS_START(2, 3) + ZEND_PARSE_PARAMETERS_START(2, 4) Z_PARAM_STRING(haystack, haystack_len) Z_PARAM_STRING(needle, needle_len) Z_PARAM_OPTIONAL Z_PARAM_LONG(loffset) + Z_PARAM_STRING(locale, locale_len) ZEND_PARSE_PARAMETERS_END(); if ( OUTSIDE_STRING(loffset, haystack_len) ) { @@ -185,7 +187,7 @@ PHP_FUNCTION(grapheme_stripos) } /* do utf16 part of the strpos */ - ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, NULL, 1 /* fIgnoreCase */, 0 /*last */ ); + ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, NULL, 1 /* fIgnoreCase */, 0, locale /*last */ ); if ( ret_pos >= 0 ) { RETURN_LONG(ret_pos); @@ -200,17 +202,19 @@ PHP_FUNCTION(grapheme_stripos) PHP_FUNCTION(grapheme_strrpos) { char *haystack, *needle; - size_t haystack_len, needle_len; + char *locale = ""; + size_t haystack_len, needle_len, locale_len = 0; zend_long loffset = 0; int32_t offset = 0; zend_long ret_pos; int is_ascii; - ZEND_PARSE_PARAMETERS_START(2, 3) + ZEND_PARSE_PARAMETERS_START(2, 4) Z_PARAM_STRING(haystack, haystack_len) Z_PARAM_STRING(needle, needle_len) Z_PARAM_OPTIONAL Z_PARAM_LONG(loffset) + Z_PARAM_STRING(locale, locale_len) ZEND_PARSE_PARAMETERS_END(); if ( OUTSIDE_STRING(loffset, haystack_len) ) { @@ -242,7 +246,7 @@ PHP_FUNCTION(grapheme_strrpos) /* else we need to continue via utf16 */ } - ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, NULL, 0 /* f_ignore_case */, 1/* last */); + ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, NULL, 0 /* f_ignore_case */, 1, locale /* last */); if ( ret_pos >= 0 ) { RETURN_LONG(ret_pos); @@ -257,18 +261,19 @@ PHP_FUNCTION(grapheme_strrpos) /* {{{ Find position of last occurrence of a string within another, ignoring case */ PHP_FUNCTION(grapheme_strripos) { - char *haystack, *needle; - size_t haystack_len, needle_len; + char *haystack, *needle, *locale = ""; + size_t haystack_len, needle_len, locale_len = 0; zend_long loffset = 0; int32_t offset = 0; zend_long ret_pos; int is_ascii; - ZEND_PARSE_PARAMETERS_START(2, 3) + ZEND_PARSE_PARAMETERS_START(2, 4) Z_PARAM_STRING(haystack, haystack_len) Z_PARAM_STRING(needle, needle_len) Z_PARAM_OPTIONAL Z_PARAM_LONG(loffset) + Z_PARAM_STRING(locale, locale_len) ZEND_PARSE_PARAMETERS_END(); if ( OUTSIDE_STRING(loffset, haystack_len) ) { @@ -309,7 +314,7 @@ PHP_FUNCTION(grapheme_strripos) /* else we need to continue via utf16 */ } - ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, NULL, 1 /* f_ignore_case */, 1 /*last */); + ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, offset, NULL, 1 /* f_ignore_case */, 1, locale /*last */); if ( ret_pos >= 0 ) { RETURN_LONG(ret_pos); @@ -324,10 +329,10 @@ PHP_FUNCTION(grapheme_strripos) /* {{{ Returns part of a string */ PHP_FUNCTION(grapheme_substr) { - char *str; + char *str, *locale = ""; zend_string *u8_sub_str; UChar *ustr; - size_t str_len; + size_t str_len, locale_len = 0; int32_t ustr_len; zend_long lstart = 0, length = 0; int32_t start = 0; @@ -339,11 +344,12 @@ PHP_FUNCTION(grapheme_substr) int32_t (*iter_func)(UBreakIterator *); bool no_length = true; - ZEND_PARSE_PARAMETERS_START(2, 3) + ZEND_PARSE_PARAMETERS_START(2, 4) Z_PARAM_STRING(str, str_len) Z_PARAM_LONG(lstart) Z_PARAM_OPTIONAL Z_PARAM_LONG_OR_NULL(length, no_length) + Z_PARAM_STRING(locale, locale_len) ZEND_PARSE_PARAMETERS_END(); if (lstart < INT32_MIN || lstart > INT32_MAX) { @@ -537,17 +543,18 @@ PHP_FUNCTION(grapheme_substr) /* {{{ strstr_common_handler */ static void strstr_common_handler(INTERNAL_FUNCTION_PARAMETERS, int f_ignore_case) { - char *haystack, *needle; + char *haystack, *needle, *locale = ""; const char *found; - size_t haystack_len, needle_len; + size_t haystack_len, needle_len, locale_len = 0; int32_t ret_pos, uchar_pos; bool part = false; - ZEND_PARSE_PARAMETERS_START(2, 3) + ZEND_PARSE_PARAMETERS_START(2, 4) Z_PARAM_STRING(haystack, haystack_len) Z_PARAM_STRING(needle, needle_len) Z_PARAM_OPTIONAL Z_PARAM_BOOL(part) + Z_PARAM_STRING(locale, locale_len) ZEND_PARSE_PARAMETERS_END(); if ( !f_ignore_case ) { @@ -574,7 +581,7 @@ static void strstr_common_handler(INTERNAL_FUNCTION_PARAMETERS, int f_ignore_cas } /* need to work in utf16 */ - ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, 0, &uchar_pos, f_ignore_case, 0 /*last */ ); + ret_pos = grapheme_strpos_utf16(haystack, haystack_len, needle, needle_len, 0, &uchar_pos, f_ignore_case, 0, locale /*last */ ); if ( ret_pos < 0 ) { RETURN_FALSE; @@ -919,14 +926,17 @@ PHP_FUNCTION(grapheme_levenshtein) zend_long cost_ins = 1; zend_long cost_rep = 1; zend_long cost_del = 1; + char *locale = ""; + size_t locale_len = 0; - ZEND_PARSE_PARAMETERS_START(2, 5) + ZEND_PARSE_PARAMETERS_START(2, 6) Z_PARAM_STR(string1) Z_PARAM_STR(string2) Z_PARAM_OPTIONAL Z_PARAM_LONG(cost_ins) Z_PARAM_LONG(cost_rep) Z_PARAM_LONG(cost_del) + Z_PARAM_STRING(locale, locale_len) ZEND_PARSE_PARAMETERS_END(); if (cost_ins <= 0 || cost_ins > UINT_MAX / 4) { @@ -1043,7 +1053,7 @@ PHP_FUNCTION(grapheme_levenshtein) RETVAL_FALSE; goto out_bi2; } - UCollator *collator = ucol_open("", &ustatus); + UCollator *collator = ucol_open(locale, &ustatus); if (U_FAILURE(ustatus)) { intl_error_set_code(NULL, ustatus); diff --git a/ext/intl/grapheme/grapheme_util.c b/ext/intl/grapheme/grapheme_util.c index d4b6c1cbfd89a..f05d18955d89c 100644 --- a/ext/intl/grapheme/grapheme_util.c +++ b/ext/intl/grapheme/grapheme_util.c @@ -94,7 +94,7 @@ void grapheme_substr_ascii(char *str, size_t str_len, int32_t f, int32_t l, char /* {{{ grapheme_strpos_utf16 - strrpos using utf16*/ -int32_t grapheme_strpos_utf16(char *haystack, size_t haystack_len, char *needle, size_t needle_len, int32_t offset, int32_t *puchar_pos, int f_ignore_case, int last) +int32_t grapheme_strpos_utf16(char *haystack, size_t haystack_len, char *needle, size_t needle_len, int32_t offset, int32_t *puchar_pos, int f_ignore_case, int last, char* locale) { UChar *uhaystack = NULL, *uneedle = NULL; int32_t uhaystack_len = 0, uneedle_len = 0, char_pos, ret_pos, offset_pos = 0; @@ -136,7 +136,7 @@ int32_t grapheme_strpos_utf16(char *haystack, size_t haystack_len, char *needle, } status = U_ZERO_ERROR; - src = usearch_open(uneedle, uneedle_len, uhaystack, uhaystack_len, "", bi, &status); + src = usearch_open(uneedle, uneedle_len, uhaystack, uhaystack_len, locale, bi, &status); STRPOS_CHECK_STATUS(status, "Error creating search object"); if(f_ignore_case) { diff --git a/ext/intl/grapheme/grapheme_util.h b/ext/intl/grapheme/grapheme_util.h index d03194621acf3..0b6e25b01a458 100644 --- a/ext/intl/grapheme/grapheme_util.h +++ b/ext/intl/grapheme/grapheme_util.h @@ -25,8 +25,8 @@ zend_long grapheme_ascii_check(const unsigned char *day, size_t len); void grapheme_substr_ascii(char *str, size_t str_len, int32_t f, int32_t l, char **sub_str, int32_t *sub_str_len); zend_long grapheme_strrpos_ascii(char *haystack, size_t haystack_len, char *needle, size_t needle_len, int32_t offset); -int32_t grapheme_strrpos_utf16(char *haystack, size_t haystack_len, char *needle, size_t needle_len, int32_t offset, int f_ignore_case); -int32_t grapheme_strpos_utf16(char *haystack, size_t haystack_len, char *needle, size_t needle_len, int32_t offset, int *puchar_pos, int f_ignore_case, int last); +int32_t grapheme_strrpos_utf16(char *haystack, size_t haystack_len, char *needle, size_t needle_len, int32_t offset, int f_ignore_case, char* locale); +int32_t grapheme_strpos_utf16(char *haystack, size_t haystack_len, char *needle, size_t needle_len, int32_t offset, int *puchar_pos, int f_ignore_case, int last, char* locale); int32_t grapheme_split_string(const UChar *text, int32_t text_length, int boundary_array[], int boundary_array_len ); diff --git a/ext/intl/php_intl.stub.php b/ext/intl/php_intl.stub.php index d2ca22c6d299a..9a8f036865cd5 100644 --- a/ext/intl/php_intl.stub.php +++ b/ext/intl/php_intl.stub.php @@ -427,23 +427,23 @@ function numfmt_get_error_message(NumberFormatter $formatter): string {} function grapheme_strlen(string $string): int|false|null {} -function grapheme_strpos(string $haystack, string $needle, int $offset = 0): int|false {} +function grapheme_strpos(string $haystack, string $needle, int $offset = 0, string $locale = ""): int|false {} -function grapheme_stripos(string $haystack, string $needle, int $offset = 0): int|false {} +function grapheme_stripos(string $haystack, string $needle, int $offset = 0, string $locale = ""): int|false {} -function grapheme_strrpos(string $haystack, string $needle, int $offset = 0): int|false {} +function grapheme_strrpos(string $haystack, string $needle, int $offset = 0, string $locale = ""): int|false {} -function grapheme_strripos(string $haystack, string $needle, int $offset = 0): int|false {} +function grapheme_strripos(string $haystack, string $needle, int $offset = 0, string $locale = ""): int|false {} -function grapheme_substr(string $string, int $offset, ?int $length = null): string|false {} +function grapheme_substr(string $string, int $offset, ?int $length = null, string $locale = ""): string|false {} -function grapheme_strstr(string $haystack, string $needle, bool $beforeNeedle = false): string|false {} +function grapheme_strstr(string $haystack, string $needle, bool $beforeNeedle = false, string $locale = ""): string|false {} -function grapheme_stristr(string $haystack, string $needle, bool $beforeNeedle = false): string|false {} +function grapheme_stristr(string $haystack, string $needle, bool $beforeNeedle = false, string $locale = ""): string|false {} function grapheme_str_split(string $string, int $length = 1): array|false {} -function grapheme_levenshtein(string $string1, string $string2, int $insertion_cost = 1, int $replacement_cost = 1, int $deletion_cost = 1): int|false {} +function grapheme_levenshtein(string $string1, string $string2, int $insertion_cost = 1, int $replacement_cost = 1, int $deletion_cost = 1, string $locale = ""): int|false {} /** @param int $next */ function grapheme_extract(string $haystack, int $size, int $type = GRAPHEME_EXTR_COUNT, int $offset = 0, &$next = null): string|false {} diff --git a/ext/intl/php_intl_arginfo.h b/ext/intl/php_intl_arginfo.h index 012e274296a50..f778e4cacf3d4 100644 --- a/ext/intl/php_intl_arginfo.h +++ b/ext/intl/php_intl_arginfo.h @@ -1,5 +1,5 @@ /* This is a generated file, edit the .stub.php file instead. - * Stub hash: 3fa507cf4ace635ae620050db9662b3979890bc1 */ + * Stub hash: d9e331c3a1ae46f8eae07ef0d39cb9990e74a0d1 */ ZEND_BEGIN_ARG_WITH_RETURN_OBJ_INFO_EX(arginfo_intlcal_create_instance, 0, 0, IntlCalendar, 1) ZEND_ARG_OBJ_TYPE_MASK(0, timezone, IntlTimeZone|DateTimeZone, MAY_BE_STRING|MAY_BE_NULL, "null") @@ -462,6 +462,7 @@ ZEND_BEGIN_ARG_WITH_RETURN_TYPE_MASK_EX(arginfo_grapheme_strpos, 0, 2, MAY_BE_LO ZEND_ARG_TYPE_INFO(0, haystack, IS_STRING, 0) ZEND_ARG_TYPE_INFO(0, needle, IS_STRING, 0) ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, offset, IS_LONG, 0, "0") + ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, locale, IS_STRING, 0, "\"\"") ZEND_END_ARG_INFO() #define arginfo_grapheme_stripos arginfo_grapheme_strpos @@ -474,12 +475,14 @@ ZEND_BEGIN_ARG_WITH_RETURN_TYPE_MASK_EX(arginfo_grapheme_substr, 0, 2, MAY_BE_ST ZEND_ARG_TYPE_INFO(0, string, IS_STRING, 0) ZEND_ARG_TYPE_INFO(0, offset, IS_LONG, 0) ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, length, IS_LONG, 1, "null") + ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, locale, IS_STRING, 0, "\"\"") ZEND_END_ARG_INFO() ZEND_BEGIN_ARG_WITH_RETURN_TYPE_MASK_EX(arginfo_grapheme_strstr, 0, 2, MAY_BE_STRING|MAY_BE_FALSE) ZEND_ARG_TYPE_INFO(0, haystack, IS_STRING, 0) ZEND_ARG_TYPE_INFO(0, needle, IS_STRING, 0) ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, beforeNeedle, _IS_BOOL, 0, "false") + ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, locale, IS_STRING, 0, "\"\"") ZEND_END_ARG_INFO() #define arginfo_grapheme_stristr arginfo_grapheme_strstr @@ -495,6 +498,7 @@ ZEND_BEGIN_ARG_WITH_RETURN_TYPE_MASK_EX(arginfo_grapheme_levenshtein, 0, 2, MAY_ ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, insertion_cost, IS_LONG, 0, "1") ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, replacement_cost, IS_LONG, 0, "1") ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, deletion_cost, IS_LONG, 0, "1") + ZEND_ARG_TYPE_INFO_WITH_DEFAULT_VALUE(0, locale, IS_STRING, 0, "\"\"") ZEND_END_ARG_INFO() ZEND_BEGIN_ARG_WITH_RETURN_TYPE_MASK_EX(arginfo_grapheme_extract, 0, 2, MAY_BE_STRING|MAY_BE_FALSE) diff --git a/ext/intl/tests/grapheme_levenshtein.phpt b/ext/intl/tests/grapheme_levenshtein.phpt index 4ff7dbb607bcd..37978f422363d 100644 --- a/ext/intl/tests/grapheme_levenshtein.phpt +++ b/ext/intl/tests/grapheme_levenshtein.phpt @@ -58,6 +58,14 @@ $nabe = '邊'; $nabe_E0100 = "邊󠄀"; var_dump(grapheme_levenshtein($nabe, $nabe_E0100)); +// variable $nabe and $nabe_E0101 is different because that is IVS. +// $nabe_E0101 is variable selector in U+908A U+E0101. +// grapheme_levenshtein can catches different only match strength is u-ks-identic for locale. +// So result is expect to 1. +$nabe = '邊'; +$nabe_E0101 = "\u{908A}\u{E0101}"; +var_dump(grapheme_levenshtein($nabe, $nabe_E0101, locale: "ja_JP-u-ks-identic")); + // combining character var_dump(grapheme_levenshtein("\u{0065}\u{0301}", "\u{00e9}")); @@ -80,6 +88,12 @@ try { } catch (ValueError $e) { echo $e->getMessage() . PHP_EOL; } + +echo "--- Invalid locales ---\n"; +var_dump(grapheme_levenshtein("abc", "ａｂｃ", locale: "defaaaaaaaaaaaaaaaaaaaaaaaaaaaa")); +var_dump(intl_get_error_code()); +var_dump(intl_get_error_message()); + ?> --EXPECTF-- --- Equal --- @@ -121,8 +135,13 @@ int(2) --- Variable selector --- int(1) int(0) +int(1) int(0) --- Corner case --- grapheme_levenshtein(): Argument #3 ($insertion_cost) must be greater than 0 and less than or equal to %d grapheme_levenshtein(): Argument #4 ($replacement_cost) must be greater than 0 and less than or equal to %d grapheme_levenshtein(): Argument #5 ($deletion_cost) must be greater than 0 and less than or equal to %d +--- Invalid locales --- +bool(false) +int(%d) +string(68) "grapheme_levenshtein(): Error on ucol_open: U_ILLEGAL_ARGUMENT_ERROR" diff --git a/ext/intl/tests/grapheme_stripos_locale_dependency.phpt b/ext/intl/tests/grapheme_stripos_locale_dependency.phpt new file mode 100644 index 0000000000000..e1a8d05e2541d --- /dev/null +++ b/ext/intl/tests/grapheme_stripos_locale_dependency.phpt @@ -0,0 +1,23 @@ +--TEST-- +grapheme_stripos() function locale dependency test +--EXTENSIONS-- +intl +--FILE-- + +--EXPECTF-- +int(0) +int(0) +bool(false) +=== Invalid locales === +bool(false) +int(%d) +string(74) "grapheme_stripos(): Error creating search object: U_ILLEGAL_ARGUMENT_ERROR" diff --git a/ext/intl/tests/grapheme_stristr_locale_dependency.phpt b/ext/intl/tests/grapheme_stristr_locale_dependency.phpt new file mode 100644 index 0000000000000..91d0e7273be0c --- /dev/null +++ b/ext/intl/tests/grapheme_stristr_locale_dependency.phpt @@ -0,0 +1,23 @@ +--TEST-- +grapheme_stristr() function locale dependency test +--EXTENSIONS-- +intl +--FILE-- + +--EXPECTF-- +string(3) "abc" +string(1) "i" +bool(false) +=== Invalid locales === +bool(false) +int(%d) +string(74) "grapheme_stristr(): Error creating search object: U_ILLEGAL_ARGUMENT_ERROR" diff --git a/ext/intl/tests/grapheme_strripos_locale_dependency.phpt b/ext/intl/tests/grapheme_strripos_locale_dependency.phpt new file mode 100644 index 0000000000000..87e87c483b7a4 --- /dev/null +++ b/ext/intl/tests/grapheme_strripos_locale_dependency.phpt @@ -0,0 +1,23 @@ +--TEST-- +grapheme_strripos() function locale dependency test +--EXTENSIONS-- +intl +--FILE-- + +--EXPECTF-- +int(0) +int(0) +bool(false) +=== Invalid locales === +bool(false) +int(%d) +string(75) "grapheme_strripos(): Error creating search object: U_ILLEGAL_ARGUMENT_ERROR"